Atmospheric Temperature vs Niño

https://www.ncei.noaa.gov/data/global-historical-climatology-network-daily/doc/GHCND_documentation.pdf

import warnings
warnings.filterwarnings("ignore")

import os
import os.path as op
import sys

import numpy as np
import pandas as pd
import seaborn as sns
import matplotlib.pyplot as plt
import df2img


sys.path.append("../../../../indicators_setup")
from ind_setup.colors import get_df_col, plotting_style
from ind_setup.tables import plot_df_table
from ind_setup.plotting_int import plot_oni_index_th
from ind_setup.plotting import plot_bar_probs_ONI

plotting_style()
from ind_setup.core import fontsize

sys.path.append("../../../functions")
from data_downloaders import GHCN, download_oni_index

Define location and variables of interest

country = 'Palau'
vars_interest = ['tmax', 'TMAX']

Get Data

update_data = False
path_data = "../../../data"

Using Koror Station

Analysis of how much the maximum and minimum temperatures over time are changing.
The analysis of the difference between these 2 variables will allow us to know how the daily variability is being modified

if update_data:
    df_country = GHCN.get_country_code(country)
    print(f'The GHCN code for {country} is {df_country["Code"].values[0]}')

    df_stations = GHCN.download_stations_info()
    df_country_stations = df_stations[df_stations['ID'].str.startswith(df_country.Code.values[0])]
    print(f'There are {df_country_stations.shape[0]} stations in {country}')

if update_data:
    GHCND_dir = 'https://www.ncei.noaa.gov/data/global-historical-climatology-network-daily/access/'
    id = 'PSW00040309' # Koror Station
    dict_min = GHCN.extract_dict_data_var(GHCND_dir, 'TMIN', df_country_stations.loc[df_country_stations['ID'] == id])[0][0]
    dict_max = GHCN.extract_dict_data_var(GHCND_dir, 'TMAX', df_country_stations.loc[df_country_stations['ID'] == id])[0][0]
    st_data = pd.concat([dict_min['data'], (dict_max['data'])], axis=1).dropna()
    st_data['TMIN'] = np.where(st_data['TMIN'] >50, np.nan, st_data['TMIN'])
    st_data['diff'] = st_data['TMAX'] - st_data['TMIN']
    st_data['TMEAN'] = (st_data['TMAX'] + st_data['TMIN'])/2
    st_data.to_pickle(op.join(path_data, 'GHCN_surface_temperature.pkl'))
else:
    st_data = pd.read_pickle(op.join(path_data, 'GHCN_surface_temperature.pkl'))

ONI index

https://origin.cpc.ncep.noaa.gov/products/analysis_monitoring/ensostuff/ONI_v5.php

p_data = 'https://psl.noaa.gov/data/correlation/oni.data'

df1 = download_oni_index(p_data)

lims = [-.5, .5]
plot_oni_index_th(df1, lims = lims)

st_data_monthly = st_data.resample('M').mean()
st_data_monthly.index = pd.DatetimeIndex(st_data_monthly.index).to_period('M').to_timestamp() + pd.offsets.MonthBegin(1)
st_data_monthly

	TMIN	TMAX	diff	TMEAN
DATE
1951-08-01	23.967742	31.045161	7.077419	27.506452
1951-09-01	23.993548	30.345161	6.351613	27.169355
1951-10-01	23.990000	30.673333	6.683333	27.331667
1951-11-01	24.509677	31.425806	6.916129	27.967742
1951-12-01	24.360000	31.023333	6.663333	27.691667
...	...	...	...	...
2024-10-01	25.846667	29.926667	4.080000	27.886667
2024-11-01	25.855172	30.382759	4.527586	28.118966
2024-12-01	25.859259	30.148148	4.288889	28.003704
2025-01-01	25.788889	30.422222	4.633333	28.105556
2025-02-01	25.641176	30.170588	4.529412	27.905882

883 rows × 4 columns

rolling_mean = 6 #months
df1['tmin'] = st_data_monthly['TMIN'].rolling(window=rolling_mean).mean()
df1['tmax'] = st_data_monthly['TMAX'].rolling(window=rolling_mean).mean()

df1['tdiff'] = df1['tmax'] - df1['tmin']
df1['tmean'] = (df1['tmax'] + df1['tmin'])/2

fig, ax = plt.subplots(figsize = [15, 6])
df1.ONI.plot(ax = ax, color = get_df_col()[0], lw = 2)

ax2 = ax.twinx()
df1.tmin.plot(ax = ax2, color = get_df_col()[1], lw = 2)
# df1.tmax.plot(ax = ax2, color = get_df_col()[1], lw = 2)
# df1.tdiff.plot(ax = ax2, color = get_df_col()[1], lw = 2)
# df1.tmean.plot(ax = ax2, color = get_df_col()[1], lw = 2)

<Axes: >

low_lim = np.nanmin(df1.tmin)

fig, ax = plt.subplots(figsize = [15, 6])
df1.tmin.plot(ax = ax, color = get_df_col()[1], lw = 2)

ax.fill_between(df1.index, low_lim, df1.tmin, where = (df1.ONI > lims[1]), color = 'lightcoral',
                 alpha = 0.7, label = f'ONI over th: {lims[1]}')
ax.fill_between(df1.index, low_lim, df1.tmin, where = (df1.ONI < lims[0]), color = 'lightblue', 
                alpha = 0.7, label = f'ONI below th: {lims[0]}')

ax.fill_between(df1.index, low_lim, df1.tmin, where = ((df1.ONI > lims[0]) & (df1.ONI < lims[1])),
                 color = 'lightgrey', alpha = 0.075)

ax.legend(fontsize=fontsize)
ax.set_title('TMIN and ONI', fontsize = fontsize)
ax.set_ylabel('TMIN [°C]', fontsize = fontsize)
ax.set_xlabel('Time', fontsize = fontsize)

Text(0.5, 0, 'Time')

low_lim = np.nanmin(df1.tmax)
fig, ax = plt.subplots(figsize = [15, 6])
df1.tmax.plot(ax = ax, color = get_df_col()[1], lw = 2)

ax.fill_between(df1.index, low_lim, df1.tmax, where = (df1.ONI > lims[1]), color = 'lightcoral',
                 alpha = 0.7, label = f'ONI over th: {lims[1]}')
ax.fill_between(df1.index, low_lim, df1.tmax, where = (df1.ONI < lims[0]), color = 'lightblue', 
                alpha = 0.7, label = f'ONI below th: {lims[0]}')

ax.fill_between(df1.index, low_lim, df1.tmax, where = ((df1.ONI > lims[0]) & (df1.ONI < lims[1])),
                 color = 'lightgrey', alpha = 0.075)

ax.legend(fontsize = fontsize)
ax.set_title('TMAX and ONI', fontsize = fontsize)
ax.set_ylabel('TMAX [°C]', fontsize = fontsize)
ax.set_xlabel('Time', fontsize = fontsize)

Text(0.5, 0, 'Time')

plt.figure(figsize=(5, 4))
sns.heatmap(df1.corr(), annot=True, cmap='coolwarm', vmin=-1, vmax=1)
plt.title('Correlation Heatmap')
plt.show()

df1.groupby(df1.index.year).max().tmean.min()

np.float64(26.77326692356587)

df1['tmean_ref'] = df1['tmean'] - df1.loc['1961':'1990'].tmean.mean()
df1['tmean_ref_min'] = df1['tmean'] - df1.groupby(df1.index.year).max().tmean.min()

df2 = df1.resample('Y').mean()
df2['oni_cat'] = np.where(df2.ONI > lims[1], 1, np.where(df2.ONI < lims[0], -1, 0))   
fig = plot_bar_probs_ONI(df2, var='tmean_ref')
fig.suptitle('Temperature Anomaly over the 1961-1990 mean', fontsize = fontsize)
plt.show()

df_format = np.round(df1.describe(), 2)

fig = plot_df_table(df_format)
df2img.save_dataframe(fig=fig, filename="getting_started.png")